European Archives of Oto-Rhino-Laryngology (2018) 275:2705–2711

https://doi.org/10.1007/s00405-018-5146-6

OTOLOGY

Association of benign paroxysmal positional vertigo with vitamin D

deficiency: a systematic review and meta-analysis
Mohammed A. AlGarni1 · Ahmad A. Mirza2   · Awwadh A. Althobaiti3 · Hanan H. Al‑Nemari4 · Lamees S. Bakhsh4

Received: 27 April 2018 / Accepted: 5 September 2018 / Published online: 9 October 2018
© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract
Background and objective  Benign paroxysmal positional vertigo (BPPV) is an idiopathic recurrent inner ear illness that
is caused most often by an imbalance in the metabolism of calcium carbonate crystals (otoconia) inside the semicircular
canals, in which the otoconia begin to circulate freely after being dislodged from the basic structure. The underlying etiol-
ogy of this imbalance has not yet been well established; however, a few recent articles have revealed that vitamin D level
abnormality in these patients might play a role. Therefore, we conducted the current systematic review analysis to explore
potential associations of vitamin D level with the occurrence as well as the recurrence of BPPV disease.
Methods  A comprehensive literature search was conducted using different databases to retrieve all of the articles that have
evaluated possible associations, irrespective of the study design. Then, we reported different vitamin D3 levels from BPPV
groups and control groups to estimate the standardized mean difference (SMD) between the BPPV and control groups. We
also calculated the effect size of each study under the random effects statistical model.
Results  Of the 703 studies that we identified, only 37 studies were found to be potential for our analysis, and of these, only
seven met our predetermined criteria. Two meta-analyses were conducted with respect to the occurrence and the recurrence
of BPPV. When the BPPV cases were compared to the controls (free of BPPV disease), there was an insignificant reduc-
tion in vitamin D level among the diseased groups (SMD = − 2.20; 95% CI − 6.66 to 2.26). In contrast, when the recurrent
BPPV groups were compared with the non-recurrent BPPV groups, the statistical analysis showed significantly lower level
of vitamin D among the recurrence BPPV groups (SMD = − 4.47; 95% CI − 7.55 to − 1.29).
Conclusion  Although a negative vitamin D imbalance has been reported among some BPPV patients, this review analysis
failed to establish a relationship between the occurrence of BPPV and low vitamin D level. However, low vitamin D level
was significantly evident among patients with recurrent episodes of BPPV.

Keywords  Benign positional paroxysmal vertigo · Meta-analysis · Systematic review · Vitamin D deficiency

Introduction

In broad terms, vertigo can be defined as an internal sensa-

tion of head movement in which a person finds himself or
herself spinning within their surroundings, while in reality,
* Ahmad A. Mirza
aamirza1@kau.edu.sa they are not. Specifically, positional vertigo refers to vertigo
that induced by head motion. Benign paroxysmal positional
1
Department of Surgery‑Otolaryngology Section, King vertigo (BPPV) has been described as recurrent positional
Saud Bin Abdulaziz University for Health Sciences, King vertigo events [1]. Vertigo, as well as other types of dizzi-
Abdulaziz Medical City, Jeddah, Saudi Arabia
ness, has been one of the top ten outpatient visit causes in
2
Department of Otolaryngology, Head and Neck Surgery, the United States for more than 75 years [2]. It has been
Faculty of Medicine in Rabigh, King Abdulaziz University,
Jeddah, Saudi Arabia estimated that dizziness causes 5.6 million clinical visits
3 annually in the USA, and that BPPV constitutes between
Department of Otolaryngology‑Head and Neck Surgery,
King Abdulaziz University, Jeddah, Saudi Arabia 17 and 49% of these visits [2–4]. Another study reported a
4 lifetime prevalence of 2.4% [5]. Moreover, BPPV is the most
King Abdulaziz University, Jeddah, Saudi Arabia

13
Vol.:(0123456789)

2706
common vestibular disease throughout the human lifespan
[5–7]. It has been estimated that BPPV creates a healthcare
burden of 2 billion dollars per year [8]. From an initialized
level, 86% of BPPV patients complain of interference with
their daily activities. Actually, some quit their primary jobs
because of the associated symptoms [8]. Aside from the eco-
nomical and functional impacts on more productive indi-
viduals, elderly individuals were found to exhibit significant
deterioration in both their psychological and physical health
[9]. BPPV is not only an idiopathic inner ear disease, but it
can commonly result from head trauma [10]. Females tend
to be more affected than males at a rate of approximately 2:1
[5]. In fact, postmenopausal women experience BPPV much
more often than other individuals [8]. This attributes mainly
to the high prevalence of osteoporosis among this particular
group, as there is a growing body of evidence relates BPPV
with osteoporosis [11, 12].
BPPV symptoms develop after otoconia (calcium carbon-
ate) crystals dislodge from the membrane inside the semi-
circular canals. Given the significant association between
BPPV and osteoporosis, the disturbance in the metabolism
of both vitamin D and calcium that found in osteoporosis
is probably the key element of the pathogenesis of BPPV.
Similar to the bone structures, the otoconia matrix and den-
sity are affected by the deposition of calcium crystals and
vitamin D level [13]. Only a few clinical studies have been
conducted to evaluate the relationship between vitamin D
level and BPPV [3, 14–19]. According to the experimen-
tal studies, vitamin D plays a role in the mineralization of
the otoconia structure by maintaining calcium level inside
the semicircular canals [14]. In addition, a disturbance of
the particularly sensitive receptors lining the membranous
semicircular canals, which are modulated by vitamin D, can
elicit this imbalance [20]. Vitamin D insufficiency corre-
lated with the severity of BPPV. In fact, vitamin D supple-
mentation may decrease the recurrent attacks of BPPV, and
interestingly, a complete remission was accomplished in a
large number of cases after trials of vitamin D supplementa-
tion [21, 22]. Up to this point, this is the first meta-analysis
to explore possible associations between BPPV and serum
vitamin D level.
Methodology
This study is registered at PROSPERO International Prospec-
tive Register of Systematic Reviews (CRD42018086628).
Data sources
This review incorporated English published articles that
examined the association between BPPV and vitamin D
level. Multiple databases and related search engines were
13
European Archives of Oto-Rhino-Laryngology (2018) 275:2705–2711
used in this comprehensive review, including PubMed/
Medline, Embase, and the Cochrane Library. The system-
atic literature review included all of the studies that had been
published through July 30, 2017. Different strategies, includ-
ing the PICO (population, intervention, comparator, and
outcomes) formulation, relevant Medical Subject Headings
(MeSH) terms, and cited references, were widely consid-
ered to build up the current systematic review. The keywords
included “vitamin D” OR “calcifediol” OR “25(OH)vitamin
D” OR “25-hydroxylvitamin D” OR “25-hydroxyvitamin D”
OR “25(OH)D” combined with “benign paroxysmal posi-
tional vertigo” OR “BPPV” OR “BPV”. Two of the authors
(M.A. and A.M.) worked independently to evaluate the titles
and abstracts of all of the studies identified to include any
potentially relevant articles in the analysis. The full texts that
were identified were then evaluated for eligibility.
Inclusion and exclusion criteria
The eligibility criteria were as follows: (1) comparative
studies conducted retrospectively or prospectively among
adult patients, (2) clear definition of cases and controls in the
case–control studies, (3) outcome clearly defined and evalu-
ated, (4) vitamin D measured using a standard method and
expressed as, or converted to, one international unit (ng/ml),
(5) the exposure of interest was the serum vitamin D3, and
(6) the outcome of interest was BPPV defined as positional
vertigo confirmed by repositioning maneuvers. Those stud-
ies that included individuals diagnosed with osteoporosis
or osteopenia, either in control and/or case groups, were
excluded from the review. Books, reviews, commentaries,
abstracts, and letters were excluded from the study.
Data extraction and quality assessment
Two of the authors (A.A. and A.M.) collected the data
according to the predefined criteria. Any discrepancies were
resolved by discussions between the authors that reached a
consensus. The data were extracted and recorded on a data
extraction form, and only relevant data were retrieved from
each study, including the authors’ names, study country,
study design, publication year, age of subjects, number of
participants, means of vitamin D level and their standard
deviations, baseline characteristics of the individuals, and
control source.
The included studies were evaluated using the Newcas-
tle–Ottawa Quality Assessment Scale (NOS) for both the
cohort and case–control studies to assess the risk of bias in
each study. Three broad perspectives using a “star system”
were evaluated: (1) selection of study groups, (2) compa-
rability of study groups, and (3) ascertainment of either
the exposure or outcome of interest for the case–control or
cohort studies, respectively. Each of these perspectives was
European Archives of Oto-Rhino-Laryngology (2018) 275:2705–2711 2707
scored from 1 to 3 for each study, and the study quality was
graded as high (7–9 stars), intermediate (4–6 stars), or poor
(1–3 stars).
Statistical analysis
A statistical analysis was conducted using the Review Man-
ager (RevMan) [Computer program]. Version 5.3. Copen-
hagen: The Nordic Cochrane Centre, The Cochrane Col-
laboration, 2014. Forest plot was tested under the fixed and
random effects models, and the random effects model was
considered if there was considerable heterogeneity across the
combined studies. The standardized mean difference (SMD)
was used as the metameter and the standard deviation (SD)
was considered in evaluating the precision and significance
of that point of estimate. If a study reported their estimate
as a median and interquartile range, we calculated the mean
and SD as described by Hozo et al. [23]. Only the funnel plot
was used in this meta-analysis to determine the publication
bias.
Results
Trial selection
The titles and/or abstracts of 703 publications were
reviewed, but only 37 relevant articles were identified.
These studies were evaluated, and 26 studies were excluded
because some were out of the scope of this review and oth-
ers were duplicates that were indexed in different databases.
As many as 11 published articles were found to be more
relevant to our hypothesis; however, three of them were not
to evaluate the vitamin D level among patients with BPPV,
but rather, to evaluate the treatment effects of vitamin D on
BPPV. Another one was excluded because the precision of
the mean of the vitamin D level was presented as a range and
percentile, given in the consideration that a more consistent
value was difficult to acquire after trying to reach the cor-
responding author (Fig. 1). All of the included studies were
published within the recent 5 years.
Description of studies
Upon extensive review of the seven studies relevant to our
hypothesis, we divided them into two sub-analyses: one
considered the BPPV cases versus the controls, while the
other comparison was generated for the BPPV recurrences
versus the non-recurrences. Four of 7 of these studies com-
paring the BPPV patients (n = 253) with the healthy controls
(n = 401) were combined in the first meta-analysis (Fig. 2).
Additionally, four of the seven studies compared the patients
with recurrent BPPV (n = 86) to controls, who experienced
a single BPPV attack (n = 275). Of the seven studies, four of
them were cohorts and the others were case–control studies.
Table 1 provides more in-depth description for each included
study.
Study findings
In the first group meta-analysis (Fig. 2), there was a remark-
able heterogeneity (I squared = 90%, p < 0.001). Under the
random effects of the meta-analysis, there was no significant
difference in the vitamin D level in the BPPV patients when
compared to the healthy controls (SMD = − 2.20; 95% CI:
− 6.66 to 2.26; Fig. 2).
In the second meta-analysis (heterogeneity measures: I
squared = 41%, p = 0.17), there was a significant reduction in
the vitamin D level among the patients with recurrent BPPV
symptoms when compared to those with single BPPV attack
(SMD = − 4.47; 95% CI: − 7.55 to − 1.29; Fig. 3), under the
random effects statistical model.
Discussion
Calcium is the main component of otoconia, the structure in
the semicircular canals which are disturbed in patients with
BPPV. Vitamin D level in patients with BPPV has been a
primary objective in recently published studies, as well as
in the current meta-analysis. Vitamin D has been related to
many diseases, including BPPV. This is the first systematic
review analysis that assess the associations of vitamin D
level with BPPV disease. Therefore, we did not limit our
search to a specified time period. Nevertheless, we found
seven studies eligible for our analysis; all were published
within the last 5 years.
Based on this review, there were conflicting results deter-
mining whether vitamin D deficiency plays a significant role
in the underlying mechanism of BPPV. After analyzing the
different studies, few of those has reliable control groups not
characterized by specific bone diseases, such as osteoporo-
sis. Four of the seven studies were analyzed, and they indi-
cated that the reduction in vitamin D level among the BPPV
patients was not significantly associated with the disease
under the random effects model. Literature indicates that
there is a decreased vitamin D level in patients with upper
respiratory tract infections (URTIs) [24, 25]. In the study
conducted by Abdullah Karats et al. [3], they recruited the
controls from hospital clinics and all suffered from upper
respiratory tract infections. Therefore, higher vitamin D
level in the BPPV group can be explained by the relatively
lower levels of vitamin D among their selected unhealthy
controls. This causes underestimation of the significance of
the combined estimate in favor of the opposite side of our
hypothesis. Despite their opposing results, they indicated
13

2708 European Archives of Oto-Rhino-Laryngology (2018) 275:2705–2711

Fig. 1  Flow chart depicting the

method of study selection

Idenficaon
Records idenfied through
database searching
(n =37)

Records aer duplicates removed

(n =16)
Screening

Records screened Records excluded

(n =16) (n = 5)

Full-text arcles assessed Full-text arcles excluded

for eligibility -out of eligibility criteria
Eligibility

(n = 11) (n = 3)
-Inconsistent measures
(n = 1)

Studies included in
qualitave synthesis
(n = 7)
Included

Studies included in
quantave synthesis
(meta-analysis)
(n = 7)

that the vitamin D level was only significantly deficient
among the women with BPPV, and more remarkably, among
those who were premenopausal according to the sub-strat-
ified analysis. According to what has been published in the
literature, osteoporosis may be the main confounder in our
analysis. Although Jeong et al. [14] selected healthy controls
from the community, he failed to eliminate osteoporosis as
a confounder in the study analysis. In other words, the sig-
nificantly deficient vitamin D level among the case group
in that study might be attributed to the higher prevalence
of osteoporosis among the group of BPPV cases. Given the
significant heterogeneity found within this meta-analysis,
this insignificant result was not conclusive.
In the second meta-analysis that investigated the differ-
ence between the recurrence and nonrecurrence of BPPV,
there was a significant difference in the vitamin D levels
between the two groups, which indicated that vitamin D
plays a role in the recurrent nature of BPPV. Consistently,
the literature indicated that the BPPV recurrence was fre-
quent among the patients with low bone marrow densities,
i.e., osteoporosis [11].
Few limitations were found in these two meta-analyses.
For instance, the different types of study design had remark-
able attribution to the heterogeneity. In addition, there was
a difference in the selection of the control groups across the
studies. Moreover, publication bias could not be excluded
in the both meta-analyses since a very few studies were
included in the funnel plots (Figs. 4, 5). We might shed light
on the fact that further well-designed case–control or cohort
studies, with healthy controls, are needed to evaluate this
hypothesis under rigorous methodology. In addition, studies
exploring the differences in the vitamin D levels between
recurrent and non-recurrent BPPV cases might provide

13
 Table 1  Summary of included studies that investigated the association between Vitamin D and BPPV
Author name Country Pub year Type of study Age (Mean ± SD) No. of subjects (case) No. of subjects Baseline characteristics between the groups
(control)

Abdullah Karatas [3] Turkey 2017 Case–control BPPV: 51.4 ± 12.2; 78 (BPPV) 78 (hospital-based -No significant difference in age and gender distribution 
Control: 48.9 ± 12.5 control) -No significant difference in mean T-scores and osteo-
porosis
Gülşah Çıkrıkçı Işık Turkey 2017 Case–control BPPV 64 (BPPV) 63 (healthy con- -No significant difference in age and gender distribution
[16] group: 56.2 ± 13.5 trol)
(female),
55.7 ± 12 (male);
Control group:
55.6 ± 14 (female),
59.6 ± 13.7 (male)
European Archives of Oto-Rhino-Laryngology (2018) 275:2705–2711

Seong-Hae Jeong Korea 2012 Case–control BPPV 100 (BPPV) 192 (community- -No significant difference in age and gender distribution
[14] group: 61.8 ± 11.6; based control) -Osteoporosis were significantly higher in the goup of
Control group: cases
60.3 ± 11.3
Hossam Sanyelbhaa Kuwait 2014 Cohort Recurrent BPPV 80 cases: 36 (recurrent 100 (healthy -No significant difference in age and gender distribution
Talaat [15] group: 48.3 ± 9.4; BPPV) control) -T-score was higher significantly in the control goup
Non-recurrent BPPV 44 (non-recurrent BPPV)
group: 47 ± 8.9;
Control group:
44.4 ± 11.2
Sinisa Maslovara Croatia 2017 Cohort 64 ± 12 31 cases: 5 (recurrent – -No significant difference in age distribution between
[17] BPPV); 26 (non-recur- the recurrent and non-recurrent groups
rent BPPV)
Bela Büki [18] Austria 2013 Cohort 67 (min: 45; max: 85) 18 cases: 4 (recurrent – -Number of females was 11 and of males was 7 
BPPV); 14 (non-recur-
rent BPPV)
Gu Il Rhim [19] Korea 2016 Cohort Recurrent BPPV 232 cases: 41 (recurrent – -The number of females was 2.68 times higher than that
group: 48.4 ± 13.9 ; BPPV); 191 (non-recur- of  males
Non-recurrent BPPV rent BPPV)
group: 50.8 ± 17

13
2709

2710 European Archives of Oto-Rhino-Laryngology (2018) 275:2705–2711

Fig. 2  Forest plot shows the mean difference (95% CI) for BPPV group vs. Control group (free of BPPV)

Fig. 3  Forest plot shows the mean difference (95% CI) for recurrent BPPV group vs. non-recurrent BPPV group

strength in determining the impact of vitamin D in BPPV

Fig. 4  Funnel plot of meta-analysis of BPPV vs. control (free of
BPPV) studies
development and progression.
The quality scale used for assessing the different studies
pointed out the intermediate-high quality of the included
studies, and since these were the only ones reported, we
chose to include all of them in the current meta-analysis,
regardless of their quality.
In conclusion, the current meta-analysis did not support
the relationship of lower vitamin D level with the occur-
rence of BPPV, but the result was not reliable due to the
remarkable heterogeneity as well as the unhealthy controls.
However, it does support, with unremarkable heterogeneity,
the fact that recurrent BPPV episodes are associated with a
low level of vitamin D.
Acknowledgements  The authors are grateful to the department of
Family and Community Medicine at King Abdulaziz University in
Jeddah, Saudi Arabia for the tremendous efforts they provided dur-
ing the period of the public health master program. In addition, the
authors extend their great thanks to Prof. Michael A. Stoto, Professor
of Health System Administration and Population Health at George-
town University and the Adjunct Professor of Biostatistics and Senior
Preparedness Fellow at Harvard T.H. Chan School of Public Health
for the outstanding and valuable lectures on systematic reviews and
meta-analyses, which helped us to build up the knowledge and skills
required to conduct such a study.

Compliance with ethical standards 

Conflict of interest  None of the authors has a conflict of interest.

Fig. 5  Funnel plot of meta-analysis of recurrent BPPV vs. non-recur-

rent BPPV studies

13
European Archives of Oto-Rhino-Laryngology (2018) 275:2705–2711 2711

References in idiopathic benign paroxysmal positional vertigo. J Neurol

260(3):832–838
15. Talaat HS, Abuhadied G, Talaat AS, Abdelaal MSS (2015) Low
1. Bhattacharyya N, Gubbels SP, Schwartz SR, Edlow JA, El-
bone mineral density and vitamin D deficiency in patients with
Kashlan H, Fife T, Holmberg JM, Mahoney K, Hollingsworth
benign positional paroxysmal vertigo. Eur Arch Otorhinolaryngol
DB, Roberts R, Seidman MD (2017) Clinical practice guideline:
272(9):2249–2253
benign paroxysmal positional vertigo (update). Otolaryngol Head
16. Işık G, Çevik Y, Emektar E, Çorbacıoğlu ŞK (2017) Analysis
Neck Surg 156(3_suppl):S1–S47
of vitamin D and calcium levels in benign paroxysmal posi-
2. Schappert SM (1992) National Ambulatory Medical Care Survey:
tional vertigo. Eurasian J Emerg Med 16(3):128–132. https​://doi.
1989 summary. Vital and health statistics. Series 13, Data from
org/10.5152/eajem​.2017.58077​
the National Health Survey, vol 110, pp 1–80
17. Maslovara S, Butkovic Soldo S, Sestak A, Milinkovic K, Rogic-
3. Karatas A, Acar Yuceant G, Yuce T, Haci C, Cebi IT, Salviz
Namacinski J, Soldo A (2017) 25 (OH) D3 levels,incidence and
M (2017) Association of benign paroxysmal positional vertigo
recurrence of different clinical forms of benig paroxysmal posi-
with osteoporosis and vitamin D deficiency: a case–controlled
tional vertigo. Braz J Otorhinolaryngol. https​://doi.org/10.1016/j.
study. J Int Adv Otol 13(2):259–265. https​://doi.org/10.5152/
bjorl​.2017.05.007
iao.2016.2640
18. Buki B, Ecker M, Junger H, Lundberg YW (2013) Vitamin D defi-
4. Hanley K, O’Dowd T, Considine N (2001) A systematic review
ciency and benign paroxysmal positioning vertigo. Med Hypoth-
of vertigo in primary care. Br J Gen Pract 51(469):666–671
eses 80(2):201–204. https​://doi.org/10.1016/j.mehy.2012.11.029
5. Neuhauser HK, Lempert T (2009) Vertigo: epidemiologic aspects.
19. Gu Il Rhim (2016) Serum vitamin D and recurrent benign par-
Semin Neurol 29(05):473–481. (© Thieme Medical Publishers)
oxysmal positional vertigo. Laryngosc Investig Otolaryngol
6. Parnes LS, Agrawal SK, Atlas J (2003) Diagnosis and manage-
1(6):150–153
ment of benign paroxysmal positional vertigo (BPPV). Can Med
20. Minasyan A, Keisala T, Zou J, Zhang Y, Toppila E, Syvälä H,
Assoc J 169(7):681–693
Lou YR, Kalueff AV, Pyykkö I, Tuohimaa P (2009) Vestibular
7. Nedzelski JM, Barber HO, McIlmoyl L (1986) Diagnoses in a
dysfunction in vitamin D receptor mutant mice. J Steroid Biochem
dizziness unit. J Otolaryngol 15(2):101–104
Mol Biol 114(3):161–166
8. Von Brevern M, Radtke A, Lezius F, Feldmann M, Ziese T, Lem-
21. Sheikhzadeh M, Lotfi Y, Mousavi A, Heidari B, Monadi M, Bakh-
pert T, Neuhauser H (2007) Epidemiology of benign paroxysmal
shi E (2016) Influence of supplemental vitamin D on intensity
positional vertigo: a population based study. J Neurol Neurosurg
of benign paroxysmal positional vertigo: a longitudinal clinical
Psychiatry 78(7):710–715
study. Casp J Intern Med 7(2):93
9. Oghalai JS, Manolidis S, Barth JL, Stewart MG, Jenkins HA
22. Talaat HS, Kabel AMH, Khaliel LH, Abuhadied G, El HAERA,
(2000) Unrecognized benign paroxysmal positional vertigo in
Talaat AS (2016) Reduction of recurrence rate of benign paroxys-
elderly patients. Otolaryngol Head Neck Surg 122(5):630–634
mal positional vertigo by treatment of severe vitamin D deficiency.
10. Motin M, Keren O, Groswasser Z, Gordon CR (2005) Benign
Auris Nasus Larynx 43(3):237–241
paroxysmal positional vertigo as the cause of dizziness in patients
23. Hozo SP, Djulbegovic B, Hozo I (2005) Estimating the mean and
after severe traumatic brain injury: diagnosis and treatment. Brain
variance from the median, range, and the size of a sample. BMC
Inj 19:693–697
Med Res Methodol 5:13
11. Yu S, Liu F, Cheng Z, Wang Q (2014) Association between osteo-
24. Ginde AA, Mansbach JM, Camargo CA Jr (2009) Association
porosis and benign paroxysmal positional vertigo: a systematic
between serum 25-hydroxyvitamin D level and upper respiratory
review. BMC Neurol 14(1):110
tract infection in the Third National Health and Nutrition Exami-
12. Parham K, Leonard G, Feinn RS, Lafreniere D, Kenny AM (2013)
nation Survey. Arch Intern Med 169:384–390
Prospective clinical investigation of the relationship between idi-
25. Sabetta JR, DePetrillo P, Cipriani RJ, Smardin J, Burns LA, Lan-
opathic benign paroxysmal positional vertigo and bone turnover.
dry ML (2010) Serum 25-hydroxyvitamin d and the incidence of
Laryngoscope 123(11):2834–2839
acute viral respiratory tract infections in healthy adults. PLoS One
13. Yang H, Zhao X, Xu Y, Wang L, He Q, Lundberg YW (2011)
5:e11088. https​://doi.org/10.1371/journ​al.pone.00110​88
Matrix recruitment and calcium sequestration for spatial specific
otoconia development. PLoS One 6:e20498
14. Jeong SH, Kim JS, Shin JW, Kim S, Lee H, Lee AY, Kim JM,
Jo H, Song J, Ghim Y (2013) Decreased serum vitamin D

13